Apparatuses and methods for generating an optical signal in an optical communication network

The present disclosure is a continuation application of International Application No. PCT/CN2020/119994, entitled “APPARATUSES AND METHODS FOR GENERATING AN OPTICAL SIGNAL IN AN OPTICAL COMMUNICATION NETWORK”, filed Oct. 9, 2020, the entirety of which is hereby incorporated by reference.

The present application relates to optical communication networks, and in particular to apparatuses and methods for mitigating nonlinear impairments in optical systems.

In optical communication networks, an optical time-domain reflectometer (OTDR) is an optical test instrument used to detect optical return loss of an optical fiber that is under testing or monitoring (also referred to as a monitored optical fiber). The OTDR transmits an optical signal including a series of optical pulses into the monitored optical fiber and detects a reflected signal due to scattering (e.g., due to Rayleigh backscattering) and/or reflection back along the monitored optical fiber. The reflected signal that is detected is used to characterize the monitored optical fiber. In some applications, the OTDR is an optical equivalent of an electronic time-domain reflectometer that measures reflections caused by changes in the impedance of an electrical cable under test. The strength of the reflected signal is measured as a function of time, and plotted as a function of the length of the monitored optical fiber or the electrical cable under test.

Thus, an OTDR with greater accuracy, measurement range, and instrument resolution is in high demand in testing and troubling shooting solutions in optical communication networks. For example, the dynamic range of the OTDR is an important parameter for evaluating the performance of the OTDR. The dynamic range is defined as the difference between the initial strength of the reflected signal (at the near end of the optical fiber being tested) and a detection noise floor (which is a sum of all the noise sources and unwanted signals). The maximum length of optical fiber that can be monitored is dependent on the dynamic range of the OTDR. In order to increase the dynamic range of the OTDR, the OTDR may generate a series of optical pulses with high power. However, using a series of optical pulses with high power may introduce different kinds of impairments to an optical signal carrying payload data that is transmitted on the monitored fiber. Introduced impairments may be nonlinear optical effects, including cross-phase modulation (XPM) and cross-polarization modulation (XPolM), which highly depend on the state of polarization (SOP) of the series of optical pulses. Such impairments degrade the performance of other optical signals carrying payload data along the optical fiber.

Some conventional methods are used to reduce the nonlinear optical effects to other optical signals transmitted on the monitored fiber. For example, one known technique is to control the power of each optical pulse generated from the OTDR to have a lower power, which may result in a smaller dynamic range of the OTDR. Another conventional method uses the OTDR in an off-line manner (e.g., by disconnecting components of an optical communication network to be tested), which may interrupt services that are being provided by the optical communication network.

Thus, it is desirable to provide an optical apparatus to detect performance of an optical communication network (e.g., including a optical fiber or optical fiber links that need to be tested, measured, or certified), while reducing nonlinear optical impairments on other optical signals carrying payload data in the optical communication network and without interrupting services provided by the optical communication network.

The present disclosure provides apparatuses and methods for generating an un-polarised optical signal. In some embodiments, the un-polarised optical signal is generated by combining a first and second series of optical pulses. Each optical pulse in the first series of optical pulses has a first SOP that is orthogonal with respect to a second SOP of each optical pulse in the second series of optical pulses. As the generated optical signal includes dual polarizations (e.g., first and second SOP that are orthogonal), the generated optical signal behaves like an un-polarised signal.

According to some example aspects, the present disclosure describes an apparatus in an optical network. The apparatus comprises at least a first and a second laser and a polarization beam combiner (PBC). The first laser is configured to generate a first series of optical pulses, and the second laser is configured to generate a second series of optical pulses. The PBC is configured to receive the first and second series of optical pulses. Each optical pulse in the first series of optical pulses is received with a first state of polarization (SOP), and each optical pulse in the second series of optical pulses is received with a second SOP. The first SOP and second SOP is orthogonal with respect to each other, and the PBC is further configured to combine the first and second series of optical pulses and to output an un-polarized optical signal.

In any of the preceding aspects/embodiments, a first frequency of the first series of optical pulses generated from the first laser has a frequency offset relative to a second frequency of the second series of optical pulses generated from the second laser.

In any of the preceding aspects/embodiments, each optical pulse in at least one of the first and second series of optical pulses has a respective rising edge and a respective falling edge. The first laser is configured to, for the optical pulse in the first series of optical pulses, adjust at least one of a rising slope time on the respective rising edge or a falling slope time on the respective falling edge to be greater than zero. The second laser is configured to, for the optical pulse in the second series of optical pulses, adjust at least one of a rising slope time on the respective rising edge or a falling slope time on the respective falling edge to be greater than zero.

In any of the preceding aspects/embodiments, the apparatus further comprises at least a first and a second optical fiber. The first optical fiber couples the first laser to the PBC, and the second optical fiber couples the second laser to the PBC. The first optical fiber is disposed such that first optical fiber has a first orientation that is physically rotated 90° with respect to a second orientation of the second optical fiber, such that the first SOP of each of the first series of optical pulses received at the PBC is orthogonal to the second SOP of each of the second series of optical pulses received at the PBC.

In any of the preceding aspects/embodiments, each optical pulse in the first series of optical pulses has at least one of a power, a pulse shape, or a timing equal to each optical pulse in the second series of optical pulses.

In any of the preceding aspects/embodiments, the apparatus further comprises an optical circulator that is configured to transmit the un-polarized optical signal into a monitored optical fiber.

In any of the preceding aspects/embodiments, the apparatus further comprises a photodetector (PD) that is configured to receive a reflected signal that is scattered back along the monitored optical fiber and to measure a power and a time delay of the reflected signal.

In any of the preceding aspects/embodiments, the apparatus includes an optical time-domain reflectometer (OTDR).

In any of the preceding aspects/embodiments, each of the first and second SOPs includes a linear polarization.

According to some example aspects, the present disclosure describes a method for generating an optical signal in an optical network. The method comprises generating a first series of optical pulses and a second series of optical pulses; receiving the first and second series of optical pulses, each optical pulse in the first series of optical pulses being received with a first state of polarization (SOP), and each optical pulse in the second series of optical pulses being received with a second SOP, the first SOP and the second SOP being orthogonal with respect to each other; and combining the first and second series of optical pulses to output the optical signal that is un-polarized.

In any of the preceding aspects/embodiments, a first frequency of the first series of optical pulses generated from the first laser has a frequency offset relative to a second frequency of the second series of optical pulses generated from the second laser.

In any of the preceding aspects/embodiments, each optical pulse in at least one of the first and second series of optical pulses has a respective rising edge and a respective falling edge. The method further comprises adjusting at least one of a rising slope time on the respective rising edge or a falling slope time on the respective falling edge to be greater than zero.

In any of the preceding aspects/embodiments, the first series of optical pulses are carried on a first optical fiber that is disposed such that the first optical fiber has a first orientation that is physically rotated 90° with respect to a second orientation of the second optical fiber, such that the first SOP of each of the first series of optical pulses is orthogonal to the second SOP of each of the second series of optical. In any of the preceding aspects/embodiments, each optical pulse in the first series of optical pulses has at least one of a power, a pulse shape, or a timing equal to each optical pulse in the second series of optical pulses.

In any of the preceding aspects/embodiments, the method further comprises transmitting the optical signal into a monitored optical fiber.

In any of the preceding aspects/embodiments, the method further comprises receiving a reflected signal that is scattered back along the monitored optical fiber; and measuring a power and a time delay of the reflected signal.

In any of the preceding aspects/embodiments, each of the first and second SOPs includes a linear polarization.

In any of the preceding aspects/embodiments, the method is implemented on an optical time-domain reflectometer (OTDR). Each of the variations or alternatives, along with some potential benefits, will be discussed in more detail below.

Like reference numerals are used throughout the Figures to denote similar elements and features. While aspects of the invention will be described in conjunction with the illustrated embodiments, it will be understood that it is not intended to limit the invention to such embodiments.

The present disclosure provides an apparatus and a method for generating an optical signal that is used to characterize (reflect or represent the physical characteristics of) a monitored optical communication network. In some embodiments, the apparatus includes at least a first and second laser and a polarization beam combiner (PBC), to generate an un-polarized optical signal. The apparatus may be used in an OTDR. As discussed in more detail below, a polarized signal is one in which the signal, modeled as a wave, constrains the waves to a particular direction. A signal may be deemed polarized, even it is not perfectly polarized, if it behaves in practice like a polarized signal. Similarly, a signal may be deemed un-polarised if it effectively behaves in practice like an un-polarised signal.

In some examples, the un-polarized optical signal is generated by combining a first series of optical pulses generated from the first laser and a second series of optical pulses generated from the second laser. Each of the first series of optical pulses has a first state of polarization (SOP), and each of the second series of optical pulses has a second SOP. The first SOP and second SOP are orthogonal with respect to each other. The orthogonality need not be perfect. SOP, though representing a physical phenomenon, may be modeled mathematically, e.g., as a vector. In some applications, each optical pulse in the first series of optical pulses has at least one of a power, a pulse shape, or a timing approximately equal to each optical pulse in the second series of optical pulses. As the generated optical signal with dual polarizations (e.g., each first SOP is orthogonal with respect to each second SOP) effectively behaves like an un-polarized signal, the nonlinear optical effects (e.g., XPM or XPolM) that closely depend on polarization of the generated optical signal may be eliminated or reduced significantly. Therefore, when the monitored optical communication network is under characterization or trouble shooting, other concurrent optical signals carrying payload data in the monitored optical communication network may be unaffected by the testing, and services provided by the monitored optical communication network is prevented from being interrupted.

illustrates an example setup using a conventional OTDR that tests or diagnoses an example monitored optical communication networkthat provides varieties of services to users. In the example of, the monitored optical communication networkincludes at least one transmitter (Tx), an optional amplifier, a wavelength-division-multiplexer (WDM), an optical fiber, and other possible components that are not shown for simplicity. The transmittergenerates at least a first optical signalthat carries payload data. The optional amplifieramplifies the first optical signal(s)and then generates at least one amplified signal. The amplified signal(s) is then multiplexed onto the optical fiberby the WDM.

When the optical communication networkneeds to be monitored or diagnosed, a conventional OTDRcan be used to test links or nodes of the optical communication networkwithout interrupting the services that are being provided to the users (e.g., without disconnecting links or nodes of the optical communication network) if the wavelength of a second optical signal generated by the conventional OTDRis out-of-band (OOB). For example, each of the first optical signal(s)may have a typical C band, including a wavelength range from 1529 nanometer (nm) to 1562 nm, as shown in. If the optical communication networkneeds to be tested or diagnosed without services being interrupted, a wavelength of the second optical signalhas to be different than or outside of the wavelength range (e.g., 1529-1562 nm) of the first optical signal(s). In the example of, the wavelength of the second optical signalis around 1510 nm, which falls outside the wavelength range 1529-1562 nm occupied by the first optical signal(s).

Although the optical communication networkas shown inprovides an example optical communication environment, this is only illustrative and is not intended to be limiting. In other examples, the optical commutation networkcan be any other optical communication network and may include other components, for example the OTDR, more than one amplifier, a plurality of optical fibers, and/or more than one WDM.

While an OTDR performs diagnosis or characterization of a monitored optical communication network, various requirements should be met to ensure greater performance of the OTDR and/or to reduce impact on transmission of the payload-carrying optical signal. For example, the dynamic range is an important factor to evaluate the performance of the OTDR. When the OTDR tests an optical fiber over a long distance, the power of a signal that is reflected from the far end of the optical fiber may be too low to be distinguishable from noise. Therefore, in order to ensure that the OTDR can be used to perform diagnosis or monitoring of an optical fiber over a longer distance (i.e., larger dynamic range), the power of the second optical signalgenerated by the OTDR may be increased significantly. However, due to Kerr nonlinear interferences between neighboring channels, the second optical signalmay cause transient nonlinear phase changes to the first optical signalby introducing different kinds of nonlinear optical impairments (e.g., XPM, or XPolM) that closely depend on polarization of the second optical signal. In particular, the higher the power of the second optical signal, the severer the nonlinear optical impairments.

What is more, when the nonlinear phase changes have been introduced to the first optical signal(such as a quadrature amplitude modulated signal) at a transmitter, a phase recovery algorithm and/or polarization recovery algorithm may be applied at a receiver in the optical communication networkto track those nonlinear phase changes. However, there can be many residual phase errors that occur at rising and/or falling edges of each recovered optical signal, and rates of the nonlinear phase changes may be too fast for performing the tracking. Thus, different kinds of errors may be introduced when the payload data is recovered at the receiver, which may result in higher bit error ratio (BER).

As such, an apparatus and a method disclosed herein are used to diagnose, characterize, and/or perform troubling-shooting of a monitored optical communication network, without requiring the monitored optical communication network to be off-line (e.g., disconnecting the monitored optical communication network to interrupt services). The apparatus may help to reduce or eliminate nonlinear optical impairments caused by polarization of optical signals generated by the apparatus, and may also have a sufficiently large dynamic range.

illustrates a block diagram of an example apparatuswhich may be used to diagnose, characterize, and/or perform troubling shooting of the monitored optical communication networkof, in accordance with example embodiments disclosed herein. The apparatusincludes at least one optical signal generator. The optical signal generatorincludes at least a first and a second laser,and a polarization beam combiner (PBC). The first lasergenerates a first series of optical pulses, and the second lasergenerates a second series of optical pulses. Each optical pulse in the first series of optical pulses has at least one of power, pulse shape, or timing approximately equal to each optical pulse in the second series of optical pulses. In some examples, the power may be within around 1 dB. The first laseris coupled to the PBCvia a first optical fiber, and a second optical fibercouples the second laserto the PBC. The PBC then combines the first and second series of optical pulses produced by the first and second laser,respectively to generate a combined optical signal, with dual polarizations. The optical signal generated from the optical signal generatoris thus an un-polarized optical signal.

In particular, each pulse of the first series of optical pulses received by the PBChas a first SOP, and each pulse of the second series of optical pulses received by the PCBhas a second SOP. The first SOP and second SOP are orthogonal with respect to each other. In some examples, each of the first and second SOPs includes a linear polarization (in which the optical pulses are substantially planar along a direction of propagation). As the first SOP and second SOP are oriented to be orthogonal with respect to each other, the combined optical signal outputted by the PBCincludes two orthogonal polarizations and effectively behaves like an un-polarized optical signal. Accordingly, nonlinear polarization-dependent optical impairments caused by the optical signal generated by the PBCmay be eliminated or mitigated significantly, because the optical signal generated by the PBCbehaves like an un-polarized optical signal. Therefore, while the apparatusis used to diagnose a monitored optical communication network that is in operation or online, the dynamic range of the apparatuscan be increased to monitor longer optical fibers or links of the monitored optical communication network without severe nonlinear optical impairments.

In some examples, the apparatusmay be an OTDR, which may further comprise an optical circulator or an optical coupler, a photodetector (PD), and a processor. The un-polarized optical signal generated from the PBCis coupled to subsequent optical fibers by the optical circulatorto test, track, or perform trouble shooting on a monitored optical communication network, for example to monitor or diagnose one or more monitored optical fibers/linksof the optical communication network. The PDdetects one or more reflected signals that is scattered (e.g., Rayleigh backscattered) or reflected back by the monitored optical fiber. The processoris coupled to the PDto receive data from the PDrepresenting the detected reflected optical signals. Intensity (e.g., power) and elapsed time (e.g., time delay) of the reflected optical signals are measured and processed by the processor. The processormay be further coupled to a display (not shown) for outputting the measurements. For example, the processormay calculate the time delay and map or convert the calculated time delay into a length of the monitored optical fiberbased on known light speed in the fiber. In some examples, as the reflected signal is weak, the measurement may be repeated multiple times by using a series of optical pulses in the un-polarized signal generated by the PBC, rather than a single optical pulse.

shows a plot of optical pulses of the un-polarized signal that is generated by the PBC. In some examples, a time intervalbetween two consecutive optical pulses may be longer than a round trip time (RTT) that is required for one optical pulse to travel the full length of the monitored optical fiberand return to the PD. A unit of the time intervalmay be millisecond. For example, a RTT for an optical fiber 100 kilometres (km) long is about 1 ms. Thus, the time intervalmay be longer than 1 ms.

As mentioned above, optical pulses that are transmitted on the optical fiber for monitoring purposes may cause undesirable nonlinear phase changes (e.g., XPM effects) as well in a payload-carrying optical signal being concurrently carried on the same optical fiber. When recovering the payload-carry optical signal at a receiver (e.g., coherent receiver), the undesirable nonlinear phase changes may be tracked at the receiver in order to ensure the process of recovering to be more accurate. However, when the undesirable nonlinear phase changes are being tracked, residual phase errors (resulting in higher BER) may occur at rising and/or falling edges on each optical pulse of a recovered optical signal because rates of the undesirable nonlinear phase changes are too fast to be tracked. Thus, it may be useful to slow the rates of the undesirable nonlinear phase changes caused by the optical pulses generated from the apparatus. Doing so may help to eliminate or mitigate XPM effects (in addition to mitigation of XPolM effects by using an un-polarized optical signal as discussed above).

Referring toagain, in some possible configurations, each of the first and second laser,may further include a respective one of a first or second control circuit() or() (generically referred to as a control circuit) that adjusts (e.g., slows down) the rising edges and/or falling edges of a respective optical pulse. For example, for each of the first and second series of the optical pulses, in order to ensure each optical pulse to have non-zero rising slope time and/or falling slope time, the first control circuit() or the second control circuit() adjusts at least one of a rising slope time on the rising edge or a falling slope time on the falling edge of the corresponding optical pulse to be greater than zero (e.g., greater than 0 second, 0 nanosecond, . . . etc.). In particular, the control circuitmay adjust the rising slope time and/or falling slope time to be sufficiently slow to help enable better tracking at the receiver. Therefore, the rates of the nonlinear phase changes caused by the optical pulses generated from the apparatusmay be reduced, which may in turn help to reduce or mitigate the residual phase errors in recovering the optical signal at the receiver. Accordingly, the BER may be reduced. In this example, the first and second control circuits(),() may be used to control internal settings (e.g., hardware, software implementation (e.g., digital control of signal), or a combination of hardware and software implementations) of a respective laser,. In other possible configurations, the first and second control circuits(),() may be components that are external to the first and second laser,within the apparatus.

In some examples, if the first and second series of optical pulses are at slightly different but close frequencies (e.g., differ by a few Hz), beating can occur between the first and second series of optical pulses (beating may make it difficult to distinguish optical pulses). Therefore, a first frequency of the first series of optical pulses may have an intentionally introduced frequency offset relative to a second frequency of the second series of optical pulses, in order to separate the first and second frequencies to avoid the beating. The intentionally introduced frequency offset should be sufficient (e.g., at least a few GHz, such as 10 GHz) to avoid the beating phenomenon. This may help to reduce the nonlinear optical impairments further.

presents an example optical signal generator, which enables the first SOP of the first series of pulses and the second SOP of the second series of pulses that are received at the PBCto be orthogonal with respect to each other. In some examples, the PBCincludes two input ports each receiving a respective series of optical pulses from a respective laser,, via a respective optical fiber,. The PBCmay be configured to (arranged or otherwise set up to be able to) combine two optical signals having two orthogonal polarizations. At each input port, only one polarization of the received signal is passed through to be combined at an optical interface. As shown in, the first series of optical pulses are generated by the first laserand received at the PBCvia the first optical fiber. The first optical fibermay be rotated about its main axis() such that the first series of optical pulses are linearly polarized in one polarization (e.g., having X polarization) when the first series of pulses are received by the PCB. Similarly, the second series of optical pulses are generated by the second laserand received at the PBCvia the second optical fiber. The second optical fibermay be rotated about its main axis() such that the second series of optical pulses are linearly polarized in a second orthogonal polarization (e.g., having Y polarization) when the second series of pulses are received by the PCB. Thus, the first optical fiber is disposed to have a first orientation, and the second optical fiber is disposed to have a second orientation, where the first orientation of the first optical fiberis physically rotated 90° with respect to the second orientation of the second optical fiber. Accordingly, the SOP of the first series of optical pulses and the SOP of the second series of optical pulses are orthogonal with respect to each other when received by the PCB. The first and second series of optical pulses can then pass through the PBCand are combined by the PBCat the optical interface. The output of the PBCis the un-polarized optical signal that is the result of this combination. Such physical rotation (e.g., 90° relative rotation) of the first and second optical fibers may help to enable the first and second series of optical pulses have SOPs orthogonal with respect to each other, when received at the PBC, without introducing extra hardware cost.

is an illustrative example that utilizes physical rotations of fibers,to enable the combined optical signal generated from the PBCto have dual orthogonal polarizations. However, these examples are not intended to be limiting. In other examples, any other possible configuration in the optical signal generatormay be applied to cause the PCBto receive the first and second series of pulses having orthogonal SOPs with respect to each other, such that the combined optical signal generated from the PBCis an un-polarised signal, with dual orthogonal polarizations. For example, other embodiments may introduce additional optical components (e.g., polarizing optical filters) into the optical signal generator.

In some possible configurations, for example, as shown in, the optical signal generatormay further include a modulator. The modulatormay be an electro-absorption modulator (EAM), which can be used to modulate an intensity (e.g., power) of each pulse generated from the PBCby applying an electric voltage. The modulatorand the electric voltage may be controlled by a control circuit. As the modulatoris a polarization-insensitive modulator, the dual orthogonal polarizations of the un-polarised signal generated from the PBCwill not be affected by the modulation.

shows an alternative example optical signal generator, which generates an un-polarized signal including two SOPs orthogonal with respect to each other. In this example, the optical signal generatorcomprises a laserand a Dual Polarization Mach Zenhder Modulator (DPMZM). As shown in, the DPMZMcomprises first and second Mach Zenhder Modulators (MZMs)(),() (generically referred to as MZM) and a PBC. Compared to the example of, although one single laseris used in the example of, the two MZMs(),() can perform modulation and polarization adjustments or rotations on respective received series of optical pulses and enable the PBCto receive two series of optical pulses that have SOPs that are orthogonal with respect to each other. For examples, a series of optical pulses generated by the laseris split into two channels where a first and second split series of optical pulses are respectively transmitted. The first MZM() is controlled (e.g., by a control circuit) to adjust the first split series of optical pulses to align along one linear polarization (e.g., having X polarization), and the second MZM() is configured to control the second split series of optical pulses to have a polarization (e.g., Y polarization) orthogonal to the first series. As the two series of optical pulses received by the PBCare adjusted to have orthogonal SOPs, and an optical signal generated by the PCBincludes the dual orthogonal SOPs, the generated optical signal is un-polarized. In some examples, at least one of the first MZM() and the second MZM() may further be controlled to introduce a frequency offset between the first and second split series of optical pulses such that the orthogonality of SOPs can be achieved for the combined optical signal generated by the PCB.

As such, even though a single laseris applied in the optical signal generator, the DPMZMcan be used in the optical signal generatorto enable the optical pulses combined at two input ports of the PBCto be at orthogonal polarizations with respect to each other, in order to generate an un-polarized optical signal. As presented in, the modulation and/polarization performed at the DPMZMmay be controlled by the control circuit.

In some examples, for each optical pulses with a respective polarization (e.g., X or Y polarization), at least one of a rising slope time on the respective rising edge or a falling slope time on the respective falling edge may be controlled or adjusted by the control circuitto be greater than zero, in order to mitigate non-linear impairments (e.g., XPM) further.

Referring tonow, which illustrates another optical signal generatorin accordance with alternative embodiments. The optical signal generatorincludes a laserand a depolarizer. The depolarizeris an optical device used to scramble the polarizations of optical pulses. In some examples, the depolarizermay output randomly polarized light whatever its input or produce pseudo-random output polarization. That is, although the lasermay generate a series of optical pulses all having the same SOP, the depolarizerreceives the series of optical pulses and scrambles the SOP of the received optical pulses to generate an un-polarized signal having randomly changing polarizations. It should be noted that, in some examples, the series of optical pulses generated by the lasermay include SOPs having linear polarization or circular polarization. In the example of, with respect to mitigating non-linear impairments (e.g., XPM), for each optical pulse generated by the laser, the optical signal generatormay further include a control circuitto control at least one of a rising slope time on the respective rising edge or a falling slope time on the respective falling edge to be greater than zero. In some other examples, the control circuitmay be an internal setting of the laserto adjust the rising and/or falling edges.

It is noted that although the optical signal generatorsillustrated inhave different configurations, the optical signal generated by each optical signal generatoris an un-polarized signal (e.g., by combining pulses having two orthogonal SOPs, or by randomizing the SOPs of pulses). Therefore, nonlinear optical impairments (XPolM and/or XPM) caused by polarization of the generated optical signal may be reduced significantly. It should be understood that the optical signal generatormay have any of the above-described configurations, and variations thereof, to enable the optical signal generatorto output an un-polarized optical signal. Other embodiments of the optical signal generatormay be used, within the scope of the present disclosure, to output an un-polarized optical signal.

Furthermore, the control circuits,, anddemonstrated as respective components external to the modulator, the DPMZM, and the laserare illustrative and are not intended for limiting. In other possible applications, the control circuits,, andmay be internal settings of the modulator, the DPMZM, and the laser, or may have any other suitable configurations.

illustrates an example methodthat may be implemented at the apparatusdescribed above to generate an optical signal that is un-polarised, in order to mitigate the nonlinear optical impairment (e.g., XPM and/or XPolM) that closely depends on polarization of the generated optical signal. The methodincludes: